Learning-related Changes Research Articles

Early Caregiver Predictability Shapes Neural Indices of Statistical Learning Later in Infancy.

Caregivers play an outsized role in shaping early life experiences and development, but we often lack mechanistic insight into how exactly caregiver behavior scaffolds the neurodevelopment of specific learning processes. Here, we capitalized on the fact that caregivers differ in how predictable their behavior is to ask if infants' early environmental input shapes their brains' later ability to learn about predictable information. As part of an ongoing longitudinal study in South Africa, we recorded naturalistic, dyadic interactions between 103 (46 females and 57 males) infants and their primary caregivers at 3-6 months of age, from which we calculated the predictability of caregivers' behavior, following caregiver vocalization and overall. When the same infants were 6-12-months-old they participated in an auditory statistical learning task during EEG. We found evidence of learning-related change in infants' neural responses to predictable information during the statistical learning task. The magnitude of statistical learning-related change in infants' EEG responses was associated with the predictability of their caregiver's vocalizations several months earlier, such that infants with more predictable caregiver vocalization patterns showed more evidence of statistical learning later in the first year of life. These results suggest that early experiences with caregiver predictability influence learning, providing support for the hypothesis that the neurodevelopment of core learning and memory systems is closely tied to infants' experiences during key developmental windows.

Millisecond-scale motor control precedes sensorimotor learning in Bengalese finches.

A key goal of the nervous system in young animals is to learn motor skills. Songbirds learn to sing as juveniles, providing a unique opportunity to identify the neural correlates of skill acquisition. Prior studies have shown that spike rate variability decreases during song acquisition, suggesting a transition from rate-based neural control to the millisecond-precise motor codes known to underlie adult vocal performance. By quantifying how the ensemble of spike patterns fired by cortical neurons (the "neural vocabulary") and the relationship between spike patterns and song acoustics (the "neural code") change during song acquisition, we quantified how vocal control changes across learning in juvenile Bengalese finches. We found that despite the expected drop in rate variability (a learning-related change in spike vocabulary), the precision of the neural code in the youngest singers is the same as in adults, with 1-2 millisecond variations in spike timing transduced into quantifiably different behaviors. In contrast, fluctuations of firing rates on longer timescales fail to affect the motor output. The consistent presence of millisecond-scale motor coding during changing levels of spike rate and behavioral variability supports the view that variability early in learning stems from deliberate motor exploration rather than imprecise motor control.

Transformations in prefrontal ensemble activity underlying rapid threat avoidance learning.

The capacity to learn cues that predict aversive outcomes, and understand how to avoid those outcomes, is critical for adaptive behavior. Naturalistic avoidance often means accessing a safe location, but whether a location is safe depends on the nature of the impending threat. These relationships must be rapidly learned if animals are to survive. The prelimbic subregion (PL) of the medial prefrontal cortex (mPFC) integrates learned associations to influence these threat avoidance strategies. Prior work has focused on the role of PL activity in avoidance behaviors that are fully established, leaving the prefrontal mechanisms that drive rapid avoidance learning poorly understood. To determine when and how these learning-related changes emerge, we recorded PL neural activity using miniscope calcium imaging as mice rapidly learned to avoid a threatening cue by accessing a safe location. Over the course of learning, we observed enhanced modulation of PL activity representing intersections of a threatening cue with safe or risky locations and movements between them. We observed rapid changes in PL population dynamics that preceded changes observable in the encoding of individual neurons. Successful avoidance could be predicted from cue-related population dynamics during early learning. Population dynamics during specific epochs of the conditioned tone period correlated with the modeled learning rates of individual animals. In contrast, changes in single-neuron encoding occurred later, once an avoidance strategy had stabilized. Together, our findings reveal the sequence of PL changes that characterize rapid threat avoidance learning.

Durability of motor learning by observing.

Information about another person's movement kinematics obtained through visual observation activates brain regions involved in motor learning. Observation-related changes in these brain areas are associated with adaptive changes to feedforward neural control of muscle activation and behavioral improvements in limb movement control. However, little is known about the stability of these observation-related effects over time. Here, we used force channel trials to probe changes in lateral force production at various time points (1 min, 10 min, 30 min, 60 min, 24 h) after participants either physically performed, or observed another individual performing upper limb reaching movements that were perturbed by novel, robot-generated forces (a velocity-dependent force-field). Observers learned to predictively generate directionally and temporally specific compensatory forces during reaching, consistent with the idea that they acquired an internal representation of the novel dynamics. Participants who physically practiced in the force-field showed adaptation that was detectable at all time points, with some decay detected after 24 h. Observation-related adaptation was less temporally stable in comparison, decaying slightly after 1 h and undetectable at 24 h. Observation induced less adaptation overall than physical practice, which could explain differences in temporal stability. Visually acquired representations of movement dynamics are retained and continue to influence behavior for at least 1 h after observation.NEW & NOTEWORTHY We used force channel probes in an upper limb force-field reaching task in humans to compare the durability of learning-related changes that occurred through visual observation to those after physical movement practice. Visually acquired representations of movement dynamics continued to influence behavior for at least 1 h after observation. Our findings point to a 1-h window during which visual observation of another person could play a role in motor learning.

The Consolidation of Newly Learned Movements Depends upon the Somatosensory Cortex in Humans.

Studies using magnetic brain stimulation indicate the involvement of somatosensory regions in the acquisition and retention of newly learned movements. Recent work found an impairment in motor memory when retention was tested shortly after the application of continuous theta-burst stimulation (cTBS) to the primary somatosensory cortex, compared with stimulation of the primary motor cortex or a control zone. This finding that the somatosensory cortex is involved in motor memory retention whereas the motor cortex is not, if confirmed, could alter our understanding of human motor learning. It would indicate that plasticity in sensory systems underlies newly learned movements, which is different than the commonly held view that adaptation learning involves updates to a motor controller. Here we test this idea. Participants were trained in a visuomotor adaptation task, with visual feedback gradually shifted. Following adaptation, cTBS was applied either to M1, S1, or an occipital cortex control area. Participants were tested for retention 24 h later. It was observed that S1 stimulation led to reduced retention of prior learning, compared with stimulation of M1 or the control area (with no significant difference between M1 and control). In a further control, cTBS was applied to S1 following training with unrotated feedback, in which no learning occurred. This had no effect on movement in the retention test indicating the effects of S1 stimulation on movement are learning specific. The findings are consistent with the S1 participation in the encoding of learning-related changes to movements and in the retention of human motor memory.

Interactions between circuit architecture and plasticity in a closed-loop cerebellar system.

Determining the sites and directions of plasticity underlying changes in neural activity and behavior is critical for understanding mechanisms of learning. Identifying such plasticity from neural recording data can be challenging due to feedback pathways that impede reasoning about cause and effect. We studied interactions between feedback, neural activity, and plasticity in the context of a closed-loop motor learning task for which there is disagreement about the loci and directions of plasticity: vestibulo-ocular reflex learning. We constructed a set of circuit models that differed in the strength of their recurrent feedback, from no feedback to very strong feedback. Despite these differences, each model successfully fit a large set of neural and behavioral data. However, the patterns of plasticity predicted by the models fundamentally differed, with the direction of plasticity at a key site changing from depression to potentiation as feedback strength increased. Guided by our analysis, we suggest how such models can be experimentally disambiguated. Our results address a long-standing debate regarding cerebellum-dependent motor learning, suggesting a reconciliation in which learning-related changes in the strength of synaptic inputs to Purkinje cells are compatible with seemingly oppositely directed changes in Purkinje cell spiking activity. More broadly, these results demonstrate how changes in neural activity over learning can appear to contradict the sign of the underlying plasticity when either internal feedback or feedback through the environment is present.

Functional changes in brain organization after de novo audiomotor learning: An fMRI investigation

To learn to talk or play a musical instrument, the brain must acquire a “mapping” of the relationship between movements and their acoustical consequences. However, it remains unclear how this type of audiomotor learning alters the functional organization of the brain, and what neural networks support consolidation of that learning. This study used functional magnetic resonance imaging (fMRI) to measure functional changes in brain organization as a result of audiomotor learning from scratch. We used a novel paradigm in which subjects learned to move a joystick in different directions in a 2D workspace to achieve speech-like acoustical targets. At the end of each movement, subjects received auditory feedback corresponding to the sound associated with the direction that they moved in. Before and after training, we used resting-state fMRI to assess learning-related changes in functional connectivity. Behavioral results show that, over the course of practice with feedback, subjects gradually produced joystick movements with fewer errors, indicative of learning. Furthermore, some of this learning was maintained to the second day, indicating subjects started forming durable performance gains. Analyses of the fMRI data are currently underway and aimed at localizing changes in neural activity that scale with behavioral indices of audiomotor learning.

The human somatosensory cortex contributes to the encoding of newly learned movements

Recent studies have indicated somatosensory cortex involvement in motor learning and retention. However, the nature of its contribution is unknown. One possibility is that the somatosensory cortex is transiently engaged during movement. Alternatively, there may be durable learning-related changes which would indicate sensory participation in the encoding of learned movements. These possibilities are dissociated by disrupting the somatosensory cortex following learning, thus targeting learning-related changes which may have occurred. If changes to the somatosensory cortex contribute to retention, which, in effect, means aspects of newly learned movements are encoded there, disruption of this area once learning is complete should lead to an impairment. Participants were trained to make movements while receiving rotated visual feedback. The primary motor cortex (M1) and the primary somatosensory cortex (S1) were targeted for continuous theta-burst stimulation, while stimulation over the occipital cortex served as a control. Retention was assessed using active movement reproduction, or recognition testing, which involved passive movements produced by a robot. Disruption of the somatosensory cortex resulted in impaired motor memory in both tests. Suppression of the motor cortex had no impact on retention as indicated by comparable retention levels in control and motor cortex conditions. The effects were learning specific. When stimulation was applied to S1 following training with unrotated feedback, movement direction, the main dependent variable, was unaltered. Thus, the somatosensory cortex is part of a circuit that contributes to retention, consistent with the idea that aspects of newly learned movements, possibly learning-updated sensory states (new sensory targets) which serve to guide movement, may be encoded there.

Src-NADH dehydrogenase subunit 2 complex and recognition memory of imprinting in domestic chicks.

Src is a non-receptor tyrosine kinase participating in a range of neuronal processes, including synaptic plasticity. We have recently shown that the amounts of total Src and its two phosphorylated forms, at tyrosine-416 (activated) and tyrosine-527 (inhibited), undergoes time-dependent, region-specific learning-related changes in the domestic chick forebrain after visual imprinting. These changes occur in the intermediate medial mesopallium (IMM), a site of memory formation for visual imprinting, but not the posterior pole of the nidopallium (PPN), a control brain region not involved in imprinting. Src interacts with mitochondrial genome-coded NADH dehydrogenase subunit 2 (NADH2), a component of mitochondrial respiratory complex I. This interaction occurs at brain excitatory synapses bearing NMDA glutamate receptors. The involvement of Src-NADH2 complexes in learning and memory is not yet explored. We show for the first time that, independently of changes in total Src or total NADH2, NADH2 bound to Src immunoprecipitated from the P2 plasma membrane-mitochondrial fraction: (i) is increased in a learning-related manner in the left IMM 1 h after the end of training; (ii), is decreased in the right IMM in a learning-related way 24 h after training. These changes occurred in the IMM but not the PPN. They are attributable to learning occurring during training rather than a predisposition to learn. Learning-related changes in Src-bound NADH2 are thus time- and region-dependent.

Short- and long-term neuroplasticity interact during the perceptual learning of concurrent speech.

Plasticity from auditory experience shapes the brain's encoding and perception of sound. However, whether such long-term plasticity alters the trajectory of short-term plasticity during speech processing has yet to be investigated. Here, we explored the neural mechanisms and interplay between short- and long-term neuroplasticity for rapid auditory perceptual learning of concurrent speech sounds in young, normal-hearing musicians and nonmusicians. Participants learned to identify double-vowel mixtures during ~ 45min training sessions recorded simultaneously with high-density electroencephalography (EEG). We analyzed frequency-following responses (FFRs) and event-related potentials (ERPs) to investigate neural correlates of learning at subcortical and cortical levels, respectively. Although both groups showed rapid perceptual learning, musicians showed faster behavioral decisions than nonmusicians overall. Learning-related changes were not apparent in brainstem FFRs. However, plasticity was highly evident in cortex, where ERPs revealed unique hemispheric asymmetries between groups suggestive of different neural strategies (musicians: right hemisphere bias; nonmusicians: left hemisphere). Source reconstruction and the early (150-200ms) time course of these effects localized learning-induced cortical plasticity to auditory-sensory brain areas. Our findings reinforce the domain-general benefits of musicianship but reveal that successful speech sound learning is driven by a critical interplay between long- and short-term mechanisms of auditory plasticity, which first emerge at a cortical level.

Dynamic Changes in Local Activity and Network Interactions among the Anterior Cingulate, Amygdala, and Cerebellum during Associative Learning.

Communication between the cerebellum and forebrain structures is necessary for motor learning and has been implicated in a variety of cognitive functions. The exact nature of cerebellar-forebrain interactions supporting behavior and cognition is not known. We examined how local and network activity support learning by simultaneously recording neural activity in the cerebellum, amygdala, and anterior cingulate cortex while male and female rats were trained in trace eyeblink conditioning. Initially, the cerebellum and forebrain signal the contingency between external stimuli through increases in theta power and synchrony. Neuronal activity driving expression of the learned response was observed in the cerebellum and became evident in the anterior cingulate and amygdala as learning progressed. Aligning neural activity to the training stimuli or learned response provided a way to differentiate between learning-related activity driven by different mechanisms. Stimulus and response-related increases in theta power and coherence were observed across all three areas throughout learning. However, increases in slow gamma power and coherence were only observed when oscillations were aligned to the cerebellum-driven learned response. Percentage of learned responses, learning-related local activity, and slow gamma communication from cerebellum to forebrain all progressively increased during training. The relatively fast frequency of slow gamma provides an ideal mechanism for the cerebellum to communicate learned temporal information to the forebrain. This cerebellar response-aligned slow gamma then provides enrichment of behavior-specific temporal information to local neuronal activity in the forebrain. These dynamic network interactions likely support a wide range of behaviors and cognitive tasks that require coordination between the forebrain and cerebellum.SIGNIFICANCE STATEMENT This study presents new evidence for how dynamic learning-related changes in single neurons and neural oscillations in a cerebellar-forebrain network support associative motor learning. The current results provide an integrated mechanism for how bidirectional communication between the cerebellum and forebrain represents important external events and internal neural drive. This bidirectional communication between the cerebellum and forebrain likely supports a wide range of behaviors and cognitive tasks that require temporal precision.

Dopamine regulates decision thresholds in human reinforcement learning in males

Dopamine fundamentally contributes to reinforcement learning, but recent accounts also suggest a contribution to specific action selection mechanisms and the regulation of response vigour. Here, we examine dopaminergic mechanisms underlying human reinforcement learning and action selection via a combined pharmacological neuroimaging approach in male human volunteers (n = 31, within-subjects; Placebo, 150 mg of the dopamine precursor L-dopa, 2 mg of the D2 receptor antagonist Haloperidol). We found little credible evidence for previously reported beneficial effects of L-dopa vs. Haloperidol on learning from gains and altered neural prediction error signals, which may be partly due to differences experimental design and/or drug dosages. Reinforcement learning drift diffusion models account for learning-related changes in accuracy and response times, and reveal consistent decision threshold reductions under both drugs, in line with the idea that lower dosages of D2 receptor antagonists increase striatal DA release via an autoreceptor-mediated feedback mechanism. These results are in line with the idea that dopamine regulates decision thresholds during reinforcement learning, and may help to bridge action selection and response vigor accounts of dopamine.

ACTIVITY IN BARREL CORTEX RELATED TO TRACE EYEBLINK CONDITIONING.

In mammals several memory systems are responsible for learning and storage of associative memory. Even apparently simple behavioral tasks, like Pavlovian conditioning, have been suggested to engage, for instance, implicit and explicit memory processes. Here we used single whisker tactile trace eyeblink conditioning (TTEBC) to investigate learning and its neuronal bases in the mouse barrel column, the primary neocortical tactile representation of one whisker. Behavioral analysis showed that conditioned responses (CR) are spatially highly restricted, they generalize from the principal whisker only to its direct neighbors. Within the respective neural representation, the principal column and its direct neighbors, spike activity showed a learning-related spike rate suppression starting during the late phase of conditioning stimulus (CS) presentation that was sustained throughout the stimulus-free trace period (Trace). Trial-by-trial analysis showed that learning-related activity was independent from the generation of eyelid movements within a trial, and set in around the steepest part of the learning curve. Optogenetic silencing of responses and their learning-related changes during CS and Trace epochs blocked CR acquisition, but not its recall after learning. Silencing during the Trace alone, which carried major parts of the learning-related changes, had no effect. In summary, we demonstrate specific barrel column spike rate plasticity during TTEBC that can be partially decoupled from the CR, the learned eye closure, a hallmark of implicit learning. Our results, thus, point to a possible role of the barrel column in contributing to other kinds of memory as well.Significance StatementAssociation learning relies on different memory systems. The Pavlovian eyeblink reflex conditioning paradigm already involves the generation of a new movement ('move the eyelid', implicit memory), and to learn the task contingencies ('know the rule of the game', explicit memory). Using this paradigm, we study the involvement of whisker-related mouse primary somatosensory cortex (S1) in different memory systems. With recording and causal stimulation techniques, we found spike plasticity in S1 during presentation of the conditioned stimulus that is causal for implicit learning. Notably, however, blocking learning-related changes during stimulus-free trace period did not impair implicit memory. We conclude that S1 activity is related to the implicit and at least one other memory system, presumably the explicit one.

Relationship between the interlimb transfer of a visuomotor learning task and interhemispheric inhibition in healthy humans.

The "interlimb transfer" phenomenon consists of improved performance of the trained and untrained contralateral limbs after unilateral motor practice. We here assessed whether a visuomotor learning task can be transferred from one hemisphere to the other, whether this occurs symmetrically, and the cortical neurophysiological correlates of this phenomenon, focusing on interhemispheric connectivity measures. We enrolled 33 healthy subjects (age range: 24-73years). Participants underwent two randomized sessions, which investigated the transfer from the dominant to the nondominant hand and vice versa. Measures of cortical and intracortical excitability and interhemispheric inhibition were assessed through transcranial magnetic stimulation before and after a visuomotor task. The execution of the visuomotor task led to an improvement in motor performance with the dominant and nondominant hands and induced a decrease in intracortical inhibition in the trained hemisphere. Participants were also able to transfer the visuomotor learned skill. The interlimb transfer, however, only occurred from the dominant to the nondominant hand and positively correlated with individual learning-related changes in interhemispheric inhibition. We here demonstrated that the "interlimb transfer" of a visuomotor task occurs asymmetrically and relates to the modulation of specific inhibitory interhemispheric connections. The study results have pathophysiological, clinical, and neuro-rehabilitative implications.

Minimum perturbation theory of deep perceptual learning.

Perceptual learning (PL) involves long-lasting improvement in perceptual tasks following extensive training and is accompanied by modified neuronal responses in sensory cortical areas in the brain. Understanding the dynamics of PL and the resultant synaptic changes is important for causally connecting PL to the observed neural plasticity. This is theoretically challenging because learning-related changes are distributed across many stages of the sensory hierarchy. In this paper, we modeled the sensory hierarchy as a deep nonlinear neural network and studied PL of fine discrimination, a common and well-studied paradigm of PL. Using tools from statistical physics, we developed a mean-field theory of the network in the limit of a large number of neurons and large number of examples. Our theory suggests that, in this thermodynamic limit, the input-output function of the network can be exactly mapped to that of a deep linear network, allowing us to characterize the space of solutions for the task. Surprisingly, we found that modifying synaptic weights in the first layer of the hierarchy is both sufficient and necessary for PL. To address the degeneracy of the space of solutions, we postulate that PL dynamics are constrained by a normative minimum perturbation (MP) principle, which favors weight matrices with minimal changes relative to their prelearning values. Interestingly, MP plasticity induces changes to weights and neural representations in all layers of the network, except for the readout weight vector. While weight changes in higher layers are not necessary for learning, they help reduce overall perturbation to the network. In addition, such plasticity can be learned simply through slow learning. We further elucidate the properties of MP changes and compare them against experimental findings. Overall, our statistical mechanics theory of PL provides mechanistic and normative understanding of several important empirical findings of PL.

Differential mechanisms of learning-related change

Differential mechanisms of learning-related change

Entorhinal cortex directs learning-related changes in CA1 representations

Learning-related changes in brain activity are thought to underlie adaptive behaviours1,2. For instance, the learning of a reward site by rodents requires the development of an over-representation of that location in the hippocampus3–6. How this learning-related change occurs remains unknown. Here we recorded hippocampal CA1 population activity as mice learned a reward location on a linear treadmill. Physiological and pharmacological evidence suggests that the adaptive over-representation required behavioural timescale synaptic plasticity (BTSP)7. BTSP is known to be driven by dendritic voltage signals that we proposed were initiated by input from entorhinal cortex layer 3 (EC3). Accordingly, the CA1 over-representation was largely removed by optogenetic inhibition of EC3 activity. Recordings from EC3 neurons revealed an activity pattern that could provide an instructive signal directing BTSP to generate the over-representation. Consistent with this function, our observations show that exposure to a second environment possessing a prominent reward-predictive cue resulted in both EC3 activity and CA1 place field density that were more elevated at the cue than at the reward. These data indicate that learning-related changes in the hippocampus are produced by synaptic plasticity directed by an instructive signal from the EC3 that seems to be specifically adapted to the behaviourally relevant features of the environment.

Selective modulation of cortical population dynamics during neuroprosthetic skill learning

Brain-machine interfaces (BMIs) provide a framework for studying how cortical population dynamics evolve over learning in a task in which the mapping between neural activity and behavior is precisely defined. Learning to control a BMI is associated with the emergence of coordinated neural dynamics in populations of neurons whose activity serves as direct input to the BMI decoder (direct subpopulation). While previous work shows differential modification of firing rate modulation in this population relative to a population whose activity was not directly input to the BMI decoder (indirect subpopulation), little is known about how learning-related changes in cortical population dynamics within these groups compare.To investigate this, we monitored both direct and indirect subpopulations as two macaque monkeys learned to control a BMI. We found that while the combined population increased coordinated neural dynamics, this increase in coordination was primarily driven by changes in the direct subpopulation. These findings suggest that motor cortex refines cortical dynamics by increasing neural variance throughout the entire population during learning, with a more pronounced coordination of firing activity in subpopulations that are causally linked to behavior.

Transfer from spatial education to verbal reasoning and prediction of transfer from learning-related neural change.

Current debate surrounds the promise of neuroscience for education, including whether learning-related neural changes can predict learning transfer better than traditional performance-based learning assessments. Longstanding debate in philosophy and psychology concerns the proposition that spatial processes underlie seemingly nonspatial/verbal reasoning (mental model theory). If so, education that fosters spatial cognition might improve verbal reasoning. Here, in a quasi-experimental design in real-world STEM classrooms, a curriculum devised to foster spatial cognition yielded transfer to improved verbal reasoning. Further indicating a spatial basis for verbal transfer, students’ spatial cognition gains predicted and mediated their reasoning improvement. Longitudinal fMRI detected learning-related changes in neural activity, connectivity, and representational similarity in spatial cognition–implicated regions. Neural changes predicted and mediated learning transfer. Ensemble modeling demonstrated better prediction of transfer from neural change than from traditional measures (tests and grades). Results support in-school “spatial education” and suggest that neural change can inform future development of transferable curricula.

Learning from Ingroup Experiences Changes Intergroup Impressions.

Humans form impressions toward individuals of their own social groups (ingroup members) and of different social groups (outgroup members). Outgroup-focused theories predict that intergroup impressions are mainly shaped by experiences with outgroup individuals, while ingroup-focused theories predict that ingroup experiences play a dominant role. Here we test predictions from these two psychological theories by estimating how intergroup impressions are dynamically shaped when people learn from both ingroup and outgroup experiences. While undergoing fMRI, male participants had identical experiences with different ingroup or outgroup members and rated their social closeness and impressions toward the ingroup and the outgroup. Behavioral results showed an initial ingroup bias in impression ratings which was significantly reduced over the course of learning, with larger effects in individuals with stronger ingroup identification. Computational learning models revealed that these changes in intergroup impressions were predicted by the weight given to ingroup prediction errors. Neurally, the individual weight for ingroup prediction errors was related to the coupling between the left inferior parietal lobule and the left anterior insula, which, in turn, predicted learning-related changes in intergroup impressions. Our findings provide computational and neural evidence for ingroup-focused theories, highlighting the importance of ingroup experiences in shaping social impressions in intergroup settings.SIGNIFICANCE STATEMENT Living in multicultural societies, humans interact with individuals of their own social groups (ingroup members) and of different social groups (outgroup members). However, little is known about how people learn from the mixture of ingroup and outgroup interactions, the most natural experiences in current societies. Here, participants had identical, intermixed experiences with different ingroup and outgroup individuals and rated their closeness and impressions toward the ingroup and the outgroup. Combining computational models and fMRI, we find that the weight given to ingroup experiences (ingroup prediction errors) is the main source of intergroup impression change, captured by changes in connectivity between the parietal lobe and insula. These findings highlight the importance of ingroup experiences in shaping intergroup impressions in complex social environments.